148 INJUEY, RECOVERY, AND DEATH 



these stimulations are brouglit about, as suggested by my 

 ion-protein theory and by the investigations of Lasareff." 



In view of the importance of these relations it seems 

 desirable to point out that the hypothesis of the writer 

 explains the mechanism which makes one proportion 

 better than others and preserves this preeminence in 

 spite of variations in concentration. 



We assume that CaCl2 accumulates in the surface to 

 a greater degree than NaCl. The increase in concentra- 

 tion of CaClj in the surface is supposed to be ten times 

 as great as the corresponding increase of NaCl, so that 

 the proportions in the surface are those given in Table XI. 

 For example, when the proportions in the solution are 

 97.56 NaCl -f 2.44 CaCla, the proportion of NaCl to CaClj 

 in the surface is as 97.56 to 24.40, which is equivalents^ to 

 80 NaCl + 20 G&Gh. 



We calculate the amount of Na4XCa by the 

 usual formula: 



j^ _ (CNaiXCa) (CzCli)' 



(CNaCl)' (CCaCh) (CXZ,) 



but since 2CNa.xca^= Czci, we may write 



y^ (CNaiXCa) (2CNaiXCa)' 

 (CNaCO* (CcaCI.) (CxZ,) 



Putting ^ = 4(10-") and Cxz2 = 0.1 we get 



10-17 = - 

 whence CNa.xca = 0.000936. 



"As previously explained it is assumed that the reaction takes place 

 in a surface which is saturated with respect to NaCl and CaCla so 

 that while one of these may be displaced by the other (in case their 

 relative proportions in the solution are altered) the total concentration 

 does not change; for convenience this concentration is taken as 100 and 

 the sum of NaCl -|- CaClj is therefore always equal to 100. 



,„— 17__ (CNaiXCa)' 

 (80) < (20) (0.1) 



